The purpose of the Biomedical Informatics Core is to provide professional expertise in informatics to enable interpretation of results generated from high throughput technologies in post-genomic processing. To achieve these results, the Biomedical Informatics Core will draw upon the resources currently being built into the Biomedical Informatics Core in the Vanderbilt-Ingram Cancer Center. Collaborations with other cores and projects, especially Biostatistics and the Emerging Technologies Cores, will also play a key role (see Table 1. below showing objectives and interactions across cores and projects). Dr. Edgerton will be the Director of the overall Biomedical Informatics Core; each objective will have its own leader: Dr. Edgerton for objectives I and 2, phenotyping tissue acquisitions and developing a standard data normalization strategy; and Dr. Aliferis for objective 3, the development of new algorithms for data analysis. The Vanderbilt University Medical Center Informatics Center under Dr. William Stead endorses the efforts of the Biomedical Informatics Core for the GI SPORE (see letter of support from Dr. Stead, Informatics Architect for Vanderbilt University and Director of the Vanderbilt University Medical Center Informatics Center, in Appendix VI). The concept of a Biomedical Informatics Core extends beyond the scope of a traditional service core and builds upon the model set by Biostatistics Cores. Collaborative efforts with experimental investigators are covered by salary and equipment support for the core. This model stems from the very nature of the work being performed. Tissue informatics at the level described here is based on translation from current methods in de-identification, ongoing development in common data elements and controlled vocabulary, and is coupled with ongoing development of ontologisms. Similarly, analysis of high throughput post-genomic experimental data is a relatively new field without a proven gold standard. Finally, there is no method in existence for directly comparing gene expression array data and tissue mass spectrometry results. The methods to address these problems are being developed as a part of the core activities. This model reflects a new paradigm for interactions that bridge between a core and a program with the intent of translating technology for the exploration of high throughput postgenomic data generated by the Emerging Technologies Cores.